The most important thing a GPS satellite broadcasts is not a location. It is the time.
Each satellite carries atomic clocks and continuously transmits signals stamped with the moment they were sent. A receiver on the ground compares the arrival times of signals from at least four satellites and solves for four unknowns: latitude, longitude, altitude — and the exact time. Position is, in a sense, a by-product of very precise timekeeping.
That by-product quietly became infrastructure. Mobile networks, power grids and financial systems use GPS timing to keep equipment synchronised. Farmers steer tractors with it. Surveyors measure land with it. Delivery fleets, ride-hailing apps and the map on your phone depend on it. None of these uses was the reason GPS was built.
How a military navigation project became a platform for thousands of businesses is partly a story about engineering. It is at least as much a story about decisions.
Listening to Sputnik.
In October 1957, days after the Soviet Union launched Sputnik, two young physicists at the Johns Hopkins Applied Physics Laboratory, William Guier and George Weiffenbach, found they could work out the satellite's orbit from the Doppler shift in its radio signal. A colleague, Frank McClure, saw the inverse: if a satellite's orbit were known precisely, the same kind of measurement could tell an observer on the ground where they were.
That insight became Transit, a US Navy system that from the 1960s gave ships and submarines periodic position fixes. It worked, but slowly, in two dimensions, and with gaps between satellite passes.
Two other programmes supplied essential pieces. At the Naval Research Laboratory, Roger Easton's Timation project put highly stable clocks into orbit and showed that precise time could be broadcast from space. The Air Force's Project 621B, run with the Aerospace Corporation — whose founding president, Ivan Getting, was an early champion of satellite navigation — developed signals based on pseudo-random noise codes that a receiver could pick out and time accurately, even at very low power.
Each group had part of the answer. They were also, as parallel government programmes often are, rivals.
The Labor Day synthesis.
In 1973 the Department of Defense pushed for a single joint system. Over the Labor Day weekend that year, a small group led by Air Force Colonel Bradford Parkinson met at the Pentagon and worked out a design that drew on all three predecessors: a constellation of satellites carrying atomic clocks, broadcasting coded signals that receivers could use to calculate position and time. The programme became NAVSTAR GPS.
The first experimental Block I satellite was launched in 1978, and ten more followed by 1985 to validate the concept. Behind them sat a great deal of less visible science: orbit determination, clock modelling, corrections for relativity, and precise models of the Earth's shape. The mathematician Gladys West, working at the Naval Surface Warfare Center in Dahlgren, Virginia, was among those whose geodetic modelling fed into the system.
In 2019 the Queen Elizabeth Prize for Engineering went to four GPS pioneers — Parkinson, James Spilker, Hugo Fruehauf and Richard Schwartz — for work spanning system design, signal structure, clocks and satellites. Even that list necessarily leaves out hundreds of others.
A promise made before the system existed.
On 1 September 1983, a Soviet interceptor shot down Korean Air Lines Flight 007 after the airliner strayed into Soviet airspace. All 269 people aboard were killed. Two weeks later, the Reagan administration announced that GPS would be made available for civil aviation once it was operational.
At the time, GPS barely existed as a working system. The announcement was a political commitment as much as a technical one, and it set a durable direction: GPS would be shared.
Sharing came with conditions. GPS broadcast a precise code reserved for authorised military users and a coarser civilian code. Under a policy called Selective Availability, the military also deliberately degraded the civilian signal, limiting accuracy to about 100 metres. Civilians could use GPS — just not as well as the military could.
Builders at the edges.
Even with that handicap, a commercial industry appeared. Companies such as Trimble, founded in 1978, built receivers for surveying, marine and aviation users. One of the first handheld receivers for consumers, released in 1989, cost around $3,000.
Then civilian engineers worked around the government's limit. Differential GPS used fixed reference stations at precisely surveyed locations: by comparing what GPS reported with what they knew to be true, the stations could broadcast corrections to nearby receivers and cancel most of the deliberate error. Surveyors went further, using carrier-phase techniques to reach centimetre-level accuracy. The ecosystem was inventing on top of the platform — and around its constraints — before the platform was even complete.
The ecosystem was inventing on top of the platform — and around its constraints — before the platform was complete.
During the 1991 Gulf War, GPS proved its military value in featureless desert terrain, and accounts from the period describe commercial receivers being pressed into service alongside military units. The constellation reached initial operational capability in December 1993 and full operational capability in 1995.
Turning off the error.
In 1998 the US government announced plans for additional civilian signals. Then, at midnight between 1 and 2 May 2000, on President Clinton's order, Selective Availability was switched off. Civilian accuracy improved roughly tenfold, overnight.
Nothing about the satellites changed that night. What changed was a policy — and with it, the economics of every product that depended on knowing where something was. Car navigation and fleet tracking became far more useful. Within a few years, requirements that mobile networks be able to locate emergency callers added another push to put GPS chips into ordinary handsets.
Nothing about the satellites changed that night. What changed was a policy — and with it, the economics of everything built on top.
The stack that grew on top.
It helps to see GPS not as one invention but as layers, each built by different people with different incentives.

The bottom two layers were paid for by the public and offered without charge. Most of the commercial value was captured further up, by companies that had no part in building the satellites. That is unusual, and it was not inevitable: the civilian signal could have been restricted, licensed or degraded indefinitely. The openness of the interface — anyone could build a receiver from the published specification — is what turned a navigation system into a platform.
Timing deserves its own mention. Because a GPS receiver recovers very precise time, GPS became a cheap, globally available reference clock. Telecommunications networks use it to synchronise base stations, power utilities to time-stamp grid measurements, and financial systems to time-stamp transactions. These are among the least visible uses of GPS, and among the most consequential.

The fine print.
Platforms create dependence. GPS signals are extremely weak by the time they reach the ground, which makes them vulnerable to jamming and spoofing — a growing concern for aviation, shipping and critical infrastructure. Many systems that rely on GPS timing have limited backup.
Single-country control prompted its own response. Russia built GLONASS, Europe built Galileo and China built BeiDou. India developed NavIC, a regional system operated by ISRO, so that positioning in and around India would not depend entirely on another nation's decisions. Modern receivers increasingly combine several systems.
And the history is often told too simply. GPS was not opened to civilians in a single moment, and it was not invented by one person. It came from three competing programmes, a long series of engineering choices and several political decisions — one of them made in response to a tragedy.
What the story teaches.
GPS is the clearest example in this series of a technology whose greatest impact lay far beyond its original purpose and its original organisation. The Department of Defense did not build a logistics company, a farm-equipment business or a ride-hailing app. It built a dependable, open, well-specified capability — and others did the rest.
Research translation does not always mean creating one company or one product. Sometimes the most valuable outcome is enabling infrastructure: a reusable capability, a dataset, a test facility or a standard that makes many future innovations possible. Such work is often unglamorous, and its returns are diffuse and slow. It is also, as GPS shows, capable of extraordinary reach.
At ARTPARK we think about this when building shared resources. Project Vaani — an open dataset of Indian speech developed with IISc and funded by Google — is one small example of that intent: a resource meant to be built on by others. It is far too early, and the scale far too different, to compare its outcomes with GPS. The point is the principle, not the parallel.
The impact of a technology can extend far beyond its original invention or organisation. A dependable platform can enable an entire ecosystem.
- U.S. government, GPS.gov — system history and the Selective Availability decision.
- The Aerospace Corporation, “Brief History of GPS.”
- William H. Guier and George C. Weiffenbach, “Genesis of Satellite Navigation,” Johns Hopkins APL Technical Digest 19, no. 1 (1998).
- Bradford W. Parkinson and James J. Spilker Jr., eds., Global Positioning System: Theory and Applications (AIAA, 1996).
- Queen Elizabeth Prize for Engineering, 2019 laureates.
- ESA eoPortal, “GPS (Global Positioning System).”
- Indian Space Research Organisation, NavIC programme information.
Sources give slightly different dates for some milestones (for example, full operational capability in 1995). We use the most widely cited dates and avoid economic-impact estimates whose methods vary. ARTPARK had no involvement in the history described here.